At the high point of their influence many years ago, they [labor unions] supplied the people who worked the phones, stuffed the envelopes and walked the precincts on behalf of the Democrats. In some states, they still were a significant force, but overall, they were on the wane. Between 1983 and 2015, union membership as a share of employed workers plunged by almost half, from 20.1 percent to 11.1 percent. Not coincidentally, the drop-off was steepest in five industrial states that voted Republican in the 2016 presidential racePercentage Change in Union Density, Selected States, 1983-2015
1983 2015 Change
Wisconsin 24.2 08.4 -15.8
Michigan 30.8 15.3 -15.5
Indiana 25.2 10.1 -15.1
Pennsylvania 27.7 13.4 -14.3
Ohio 25.3 12.4 -12.9
Source: Barry T. Hirsch and David A. Macpherson, “State Union Membership Density 1964-2015,” http://unionstats.gsu.edu/State_Union_Membership_Density_1964-2015.xlsx; Barry T. Hirsch, David A. Macpherson, and Wayne G. Vroman, “Estimates of Union Density by State,” Monthly Labor Review 124, No. 7, July 2001, http://unionstats.gsu.edu/MLR_7-01_StateUnionDensity.pdf
A 2018 APSA paper looks at a related topic and its findings seem consistent with our observation. See "Demobilizing Democrats and Labor Unions: Political Effects of Right to Work Laws," by James Feigenbaum, Alexander Warren Hertel-Fernandez, and Vanessa Williamson. The abstract:
Labor unions play a central role in the Democratic party coalition, providing candidates with voters, volunteers, and contributions and lobbying government. Has the recent decline of organized labor hurt Democrats? We use the enactment of right-to-work laws—which weaken unions by removing agency shop protections—to estimate the effect of unions on politics from 1980 to 2016. Comparing counties on either side of a state and right-to-work border to causally identify the effects of the state laws, we find right-to-work laws reduce Democratic presidential vote shares by 4 to 6 percentage points. We find similar effects in US Senate, US House, and gubernatorial races, as well as state legislative control. Turnout is also 2 to 3 percentage points lower in right-to-work counties after those laws pass. We next explore mechanisms behind these effects, finding that right-to-work laws dampen organized labor campaign contributions to Democrats and that likely Democratic voters are less likely to be contacted to vote in right-to-work states. The weakening of unions also has large downstream effects both on who runs for office and state legislative policy. Fewer working class candidates serve in state legislatures and Congress and state policy moves in a more conservative direction following the passage of right-to-work laws.They identify a key mechanism by which such laws affect turnout.
We find that RTW laws are associated with a reduction in the probability that non-professional workers—but not professional workers—report get-out-the-vote contact during the campaign. Table 4 presents the results of this analysis, with a binary indicator for GOTV contact during the last campaign as the outcome. In the model with individual controls, we find that RTW laws reduce the probability that a non-professional worker reported GOTV contact by 11 percentage points but had no discernible effect on professional and managerial workersThey mention something that we discuss in the book, namely that deliberate political strategy is at work.
Aside from the theoretical contributions of the paper, our results also have bearing on current debates in U.S. politics. The anti-tax political activist Grover Norquist recently declared that while President Donald J. Trump may be historically unpopular, the GOP could still “win big” in 2020.41 The secret to the Republican party’s long-term success, Norquist argued, involved state level initiatives to weaken the power of labor unions. As Norquist explained it, if union reforms cutting the power of labor unions to recruit and retain members—like RTW laws—“are enacted in a dozen more states, the modern Democratic Party will cease to be a competitive power in American politics.” A weaker labor movement, Norquist reasoned, would not just have economic consequences. It would also have significant political repercussions, meaning that Democrats would
have substantially less of a grassroots presence on the ground during elections and less money to invest in politics